یادگیری ماشین

Open Access Article

1 - Optimized Modeling for satisfaction in the relationship between a physician and patient based on machine learnin methods
Fatemeh Saghafi mojtaba shadmehr Zainabolhoda Heshmati Hadi Veisi

Health has always been one of the most important concerns of human. The goal in this research is to know what factors cause and affect patient satisfaction in the relationship between a physician and patient. Since this relationship is a form of healthcare service, the More

Health has always been one of the most important concerns of human. The goal in this research is to know what factors cause and affect patient satisfaction in the relationship between a physician and patient. Since this relationship is a form of healthcare service, the SERVQUAL service quality assessment method has been used as a framework. However the questions have been reviewed based on the previous literature and the experts’ views, leading to a questionnaire designed for the healthcare domain. Data collection has been performed using the questionnaires on subjects selected amongst clients of Rhinoplasty Centers in Tehran. To analyze the data, three machine-learning approaches have been implemented namely Decision Tree, Support Vector Machine and Artificial Neural Networks. A number of possible factors affecting the patient-physician relationship have been used as input and patient satisfaction has been taken as output. Comparing the results of these three methods, Artificial Neural Networks method is shown to have better performance, which has therefore been used for prioritizing the effective factors in this relationship. The results indicate that reaching the information which the patient expects their physician to give is the most effective characteristic in patient satisfaction. The rank of gained features were compared with similar researches. The outcome was very similar and approved the results. Manuscript profile

Open Access Article

2 - A new algorithm based on ensemble learning for learning to rank in information retrieval
Azadeh Shakery elham ghanbari

Learning to rank refers to machine learning techniques for training a model in a ranking task. Learning to rank has been shown to be useful in many applications of information retrieval, natural language processing, and data mining. Learning to rank can be described by More

Learning to rank refers to machine learning techniques for training a model in a ranking task. Learning to rank has been shown to be useful in many applications of information retrieval, natural language processing, and data mining. Learning to rank can be described by two systems: a learning system and a ranking system. The learning system takes training data as input and constructs a ranking model. The ranking system then makes use of the learned ranking model for ranking prediction. In this paper, a new learning algorithm based on ensemble learning for learning ranking models in information retrieval is proposed. This algorithm iteratively constructs weak learners using a fraction of the training data whose weight distribution is determined based on previous weak learners. The proposed algorithm combines the weak rankers to achieve the final ranking model. This algorithm constructs a ranking model on a fraction of the training data to increase the accuracy and reduce the learning time. Experimental results based on Letor.3 benchmark dataset shows that the proposed algorithm significantly outperforms other ensemble learning algorithms. Manuscript profile

Open Access Article

3 - Survey different aspects of the problem phishing website detection and Review to existing Methods
nafise langari

One of the latest security threats in cyberspace to steal personal and financial information is created by phisher. Due to there Are various methods to detect phishing and also there is not an up-date comprehensive study on the issue, the authors Motivated to review and More

One of the latest security threats in cyberspace to steal personal and financial information is created by phisher. Due to there Are various methods to detect phishing and also there is not an up-date comprehensive study on the issue, the authors Motivated to review and analysis the proposed phishing detection methods in five categories such as: anti-phishing tools Based, data mining based, heuristic based, meta-heuristic based and machine learning based methods. The advantages and Disadvantages of each method are extracted from the current review and comparison. The outlines of this study can be suitable to identify the probability gaps in phishing detection problems for feature researches. Manuscript profile

Open Access Article

4 - Optimum modeling of patient satisfaction with the doctor based on machine learning methods
- شادمهر Zainabolhoda Heshmati Fatemeh saghafi Hadi Veisi

The patient-centered approach in the field of health has recently been proposed in the field of the medical system of our country, but until now there is no published scientific research on the factors of patient satisfaction with doctors. The present article aims to co More

The patient-centered approach in the field of health has recently been proposed in the field of the medical system of our country, but until now there is no published scientific research on the factors of patient satisfaction with doctors. The present article aims to cover the stated gap with a scientific evaluation based on the real information obtained from the field study. A questionnaire was designed for the health sector and was approved by the opinion of experts. In order to get the opinions of patients, a questionnaire was distributed among 500 people who underwent rhinoplasty in Tehran, and 395 questionnaires were collected. Three methods of decision tree, support vector machine and neural networks were used for data analysis. The analysis of the results according to the accuracy criteria showed that the most efficient method, in priority, the importance of the factors affecting the patient's satisfaction; Neural network method. The results of the analysis with this method indicate that the most effective feature in the patient's satisfaction with the doctor is the information that the patient expects the doctor to provide. The results of ranking factors in comparison with other studies that only used statistical methods for analysis showed that the results were relatively similar and confirmed each other. But the strengths of the neural network method in modeling is the strength of this method compared to the mentioned studies. Manuscript profile

Open Access Article

5 - Detecting Human Activities Based on Motion Sensors in IOT Using Deep Learning
Abbas Mirzaei fatemeh faraji

Control of areas and locations and motion sensors in the Internet of Things requires continuous control to detect human activities in different situations, which is an important challenge, including manpower and human error. Permanent human control of IoT motion sensors More

Control of areas and locations and motion sensors in the Internet of Things requires continuous control to detect human activities in different situations, which is an important challenge, including manpower and human error. Permanent human control of IoT motion sensors also seems impossible. The IoT is more than just a simple connection between devices and systems. IoT information sensors and systems help companies get a better view of system performance. This study presents a method based on deep learning and a 30-layer DNN neural network for detecting human activity on the Fordham University Activity Diagnostic Data Set. The data set contains more than 1 million lines in six classes to detect IoT activity. The proposed model had almost 90% and an error rate of 0.22 in the evaluation criteria, which indicates the good performance of deep learning in activity recognition. Manuscript profile

Open Access Article

6 - SQ-PUF: A Resistant PUF-Based Authentication Protocol against Machine-Learning Attack
Abolfazl Sajadi Bijan Alizadeh

Physically unclonable functions (PUFs) provide hardware to generate a unique challenge-response pattern for authentication and encryption purposes. An essential feature of these circuits is their unpredictability, meaning that an adversary cannot sufficiently predict fu More

Physically unclonable functions (PUFs) provide hardware to generate a unique challenge-response pattern for authentication and encryption purposes. An essential feature of these circuits is their unpredictability, meaning that an adversary cannot sufficiently predict future responses from previous observations. However, machine learning algorithms have been demonstrated to be a severe threat to PUFs since they are capable of accurately modeling their behavior. In this work, we analyze PUF security threats and propose a PUF-based authentication mechanism called SQ-PUF, which can provide good resistance to machine learning attacks. In order to make it harder to simulate or predict, we obfuscated the correlation between challenge-response pairs. Experimental results show that, unlike existing PUFs, even with a large data set, the SQ-PUF model cannot be successfully attacked with a maximum prediction accuracy of 53%, indicating that this model is unpredictable. In addition, the uniformity in this model remains almost the same as the ideal value in A-PUF. Manuscript profile

Open Access Article

7 - Machine Learning-Based Security Resource Allocation for Defending against Attacks in the Internet of Things
Nasim Navaei Vesal Hakami

Nowadays, the Internet of Things (IoT) has become the focus of security attacks due to the limitation of processing resources, heterogeneity, energy limitation in objects, and the lack of a single standard for implementing security mechanisms. In this article, a solutio More

Nowadays, the Internet of Things (IoT) has become the focus of security attacks due to the limitation of processing resources, heterogeneity, energy limitation in objects, and the lack of a single standard for implementing security mechanisms. In this article, a solution will be presented for the problem of security resources allocating to deal with attacks in the Internet of Things. Security Resource Allocation (SRA) problem in the IoT networks refers to the placement of the security resources in the IoT infrastructure. To solve this problem, it is mandatory to consider the dynamic nature of the communication environments and the uncertainty of the attackers' actions. In the traditional approaches for solving the SRA, the attacker works over based on his assumptions about the system conditions. Meanwhile, the defender collects the system's information with prior knowledge of the attacker's behavior and the targeted nodes. Unlike the mentioned traditional approaches, this research has adopted a realistic approach for the Dynamic Security Resources Allocation in the IoT to battle attackers with unknown behavior. In the stated problem, since there is a need to decide on deploying several security resources during the learning periods, the state space of the strategies is expressed in the combinatorial form. Also, the SRAIoT problem is defined as a combinatorial-adversarial multi-armed bandit problem. Since switching in the security resources has a high cost, in real scenarios, this cost is included in the utility function of the problem. Thus, the proposed framework considers the switching cost and the earned reward. The simulation results show a faster convergence of the weak regret criterion of the proposed algorithms than the basic combinatorial algorithm. In addition, in order to simulate the IoT network in a realistic context, the attack scenario has been simulated using the Cooja simulator. Manuscript profile

Open Access Article

8 - Intrusion Detection Based on Cooperation on the Permissioned Blockchain Platform in the Internet of Things Using Machine Learning
Mohammad Mahdi Abdian majid ghayori Seyed Ahmad Eftekhari

Intrusion detection systems seek to realize several objectives, such as increasing the true detection rate, reducing the detection time, reducing the computational load, and preserving the resulting logs in such a way that they cannot be manipulated or deleted by unauth More

Intrusion detection systems seek to realize several objectives, such as increasing the true detection rate, reducing the detection time, reducing the computational load, and preserving the resulting logs in such a way that they cannot be manipulated or deleted by unauthorized people. Therefore, this study seeks to solve the challenges by benefiting from the advantages of blockchain technology, its durability, and relying on IDS architecture based on multi-node cooperation. The proposed model is an intrusion detection engine based on the decision tree algorithm implemented in the nodes of the architecture. The architecture consists of several connected nodes on the blockchain platform. The resulting model and logs are stored on the blockchain platform and cannot be manipulated. In addition to the benefits of using blockchain, reduced occupied memory, the speed, and time of transactions are also improved by blockchain. In this research, several evaluation models have been designed for single-node and multi-node architectures on the blockchain platform. Finally, proof of architecture, possible threats to architecture, and defensive ways are explained. The most important advantages of the proposed scheme are the elimination of the single point of failure, maintaining trust between nodes, and ensuring the integrity of the model, and discovered logs. Manuscript profile

Open Access Article

9 - A New Hybrid Method Based on Intelligent Algorithms for Intrusion Detection in SDN-IoT
Zakaria Raeisi Fazlloah Adibnia Mahdi Yazdian

In recent years, the use of Internet of Things in societies has grown widely. On the other hand, a new technology called Software Defined Networks has been proposed to solve the challenges of the Internet of Things. The security problems in these Software Defined Networ More

In recent years, the use of Internet of Things in societies has grown widely. On the other hand, a new technology called Software Defined Networks has been proposed to solve the challenges of the Internet of Things. The security problems in these Software Defined Networks and the Internet of Things have made SDN-IoT security one of the most important concerns. On the other hand, the use of intelligent algorithms has been an opportunity that these algorithms have been able to make significant progress in various cases such as image processing and disease diagnosis. Of course, intrusion detection systems for SDN-IoT environment still face the problem of high false alarm rate and low accuracy. In this article, a new hybrid method based on intelligent algorithms is proposed. The proposed method integrates the monitoring algorithms of frequent return gate and unsupervised k-means classifier in order to obtain suitable results in the field of intrusion detection. The simulation results show that the proposed method, by using the advantages of each of the integrated algorithms and covering each other's disadvantages, has more accuracy and a lower false alarm rate than other methods such as the Hamza method. Also, the proposed method has been able to reduce the false alarm rate to 1.1% and maintain the accuracy at around 99%. Manuscript profile

Open Access Article

10 - Design and Collection of Speech Data as the First Step of Localization the Intelligent Diagnosis of Autism in Iranian Children
Maryam Alizadeh Shima tabibian

Autism Spectrum Disorder is a type of disorder in which, the patients suffer from a developmental disorder that manifests itself by symptoms such as inability to social communication. Thus, the most apparent sign of autism is a speech disorder. The first part of this pa More

Autism Spectrum Disorder is a type of disorder in which, the patients suffer from a developmental disorder that manifests itself by symptoms such as inability to social communication. Thus, the most apparent sign of autism is a speech disorder. The first part of this paper reviews research studies conducted to automatically diagnose autism based on speech processing methods. According to our review, the main speech processing approaches for diagnosing autism can be divided into two groups. The first group detects individuals with autism by processing their answers or feelings in response to questions or stories. The second group distinguishes people with autism from healthy people because of the accuracy of recognizing their spoken utterances based on automatic speech recognition systems. Despite much research being conducted outside Iran, few studies have been conducted in Iran. The most important reason for this is the lack of rich data that meet the needs of autism diagnosis based on the speech processing of suspected people. In the second part of the paper, we discuss the process of designing, collecting, and evaluating a speaker-independent dataset for autism diagnosis in Iranian children as the first step in the localization of the mentioned field. Manuscript profile

Open Access Article

11 - Combination of Instance Selection and Data Augmentation Techniques for Imbalanced Data Classification
Parastoo Mohaghegh Samira Noferesti Mehri Rajaei

Mohaghegh, S. Noferesti*, and M. Rajaei Abstract: In the era of big data, automatic data analysis techniques such as data mining have been widely used for decision-making and have become very effective. Among data mining techniques, classification is a common method fo More

Mohaghegh, S. Noferesti*, and M. Rajaei Abstract: In the era of big data, automatic data analysis techniques such as data mining have been widely used for decision-making and have become very effective. Among data mining techniques, classification is a common method for decision making and prediction. Classification algorithms usually work well on balanced datasets. However, one of the challenges of the classification algorithms is how to correctly predicting the label of new samples based on learning on imbalanced datasets. In this type of dataset, the heterogeneous distribution of the data in different classes causes examples of the minority class to be ignored in the learning process, while this class is more important in some prediction problems. To deal with this issue, in this paper, an efficient method for balancing the imbalanced dataset is presented, which improves the accuracy of the machine learning algorithms to correct prediction of the class label of new samples. According to the evaluations, the proposed method has a better performance compared to other methods based on two common criteria in evaluating the classification of imbalanced datasets, namely "Balanced Accuracy" and "Specificity". Manuscript profile

Open Access Article

12 - Application identification through intelligent traffic classification
Shaghayegh Naderi

Traffic classification and analysis is one of the big challenges in the field of data mining and machine learning, which plays an important role in providing security, quality assurance and network management. Today, a large amount of transmission traffic in the network More

Traffic classification and analysis is one of the big challenges in the field of data mining and machine learning, which plays an important role in providing security, quality assurance and network management. Today, a large amount of transmission traffic in the network is encrypted by secure communication protocols such as HTTPS. Encrypted traffic reduces the possibility of monitoring and detecting suspicious and malicious traffic in communication infrastructures (instead of increased security and privacy of the user) and its classification is a difficult task without decoding network communications, because the payload information is lost, and only the header information (which is encrypted too in new versions of network communication protocols such as TLS1.03) is accessible. Therefore, the old approaches of traffic analysis, such as various methods based on port and payload, have lost their efficiency, and new approaches based on artificial intelligence and machine learning are used in cryptographic traffic analysis. In this article, after reviewing the traffic analysis methods, an operational architectural framework for intelligent traffic analysis and classification has been designed. Then, an intelligent model for Traffic Classification and Application Identification is presented and evaluated using machine learning methods on Kaggle141. The obtained results show that the random forest model, in addition to high interpretability compared to deep learning methods, has been able to provide high accuracy in traffic classification compared to other machine learning methods. Finally, tips and suggestions about using machine learning methods in the operational field of traffic classification have been provided. Manuscript profile

Open Access Article

13 - Assessment of Spatial and temporal changes in land use using remote sensing (case study: Jayransoo rangeland, North Khorasan)
Mohabat Nadaf Reza Omidipour Hossein Sobhani

10.61186/jert.43616.8.14.131

Awareness of changes process, as well as the proper management of land use in natural ecosystems, is of great importance in conservation natural resources. In this regard, the use of remote sensing has become a common approach due to the provision an extent spatial a More

Awareness of changes process, as well as the proper management of land use in natural ecosystems, is of great importance in conservation natural resources. In this regard, the use of remote sensing has become a common approach due to the provision an extent spatial and temporal information. In this research, in order to land use mapping, first, the accuracy of three common methods of pixel-based (maximum likelihood), machine learning (support vector machine) and object-oriented methods were compared. Then, the spatial and temporal changes of land use in a period of 26 years (1997-2023) assessed using six Landsat satellite imagery. The accuracy of image classification methods was evaluated using Kappa coefficient and overall accuracy indices and the change trend was evaluated using crosstab and spatial evaluation methods. Based on the results, the support vector machine method had the highest kappa coefficient (0.71 to 0.98) and overall accuracy (86 to 99%) for all studied courses. According to the results, poor rangeland had a decreasing trend, and the land uses of very poor rangeland, bare soil, and rainfed agriculture had increasing trends. The area of poor rangeland decreased from 962 hectares (44.36%) in 1997 to 489 hectares (22.57%) in 2023, while very poor rangeland increased from 1138 hectares (52.48%) to 1606 hectares (74.05 percent) in the same period. The results of this research indicated that the trend of land use changes in Jayransoo rangeland is towards the destruction of rangelands and with the passage of time this trend is intensifying. Also, based on the results obtained from this research, it is suggested to use machine learning based classification method to prepare land use mapping in future research. Manuscript profile

Open Access Article

14 - Determining the Location of Lightning Strike Using Electromagnetic Time Reversal (EMTR) Method and Machine Learning
Abbas Hamedooni Asli m.h. m.

Determining the location of lightning strikes (LLS) is one of today's challenges in various fields, especially in the fields of electricity and electronics. To determine the location of the lightning strike, classical methods were used previously; however, the use of More

Determining the location of lightning strikes (LLS) is one of today's challenges in various fields, especially in the fields of electricity and electronics. To determine the location of the lightning strike, classical methods were used previously; however, the use of electromagnetic time reversal (EMTR) method has also become popular recently. According to the calculation of the complete waveform of the field using the EMTR method, the accuracy in determining the location of the lightning strike has significantly increased compared to the traditional methods. In the electromagnetic time reversal method with the help of finite difference time domain (FDTD), the transient electromagnetic field produced by the lightning channel is calculated first. After the time reversal of the wave, it is re-emitted from the sensor or sensors to its source and again with the help of FDTD, The re-emission electromagnetic field in the desired environment is calculated. With the electromagnetic field of the environment, using criteria (such as maximum field amplitude, maximum energy and entropy, etc.), the location of the lightning strike is determined. In traditional methods, it is quite difficult to determine the uniqueness of the final response in environments with different characteristics, and the use of at least three sensors is mandatory. In this paper, to overcome these limitations, a method based on the combination of machine learning and three-dimensional EMTR (3D-FDTD) is proposed to determine the lightning strike location. First, the three-dimensional time domain finite difference method is used to calculate the electromagnetic field of the environment and using EMTR, the back-diffusion electromagnetic field (again with the help of 3D-FDTD) is calculated in the entire environment. In this way, the necessary data for the production of RGB image profiles is prepared. Then VGG19, a pre-trained convolutional neural network (CNN), is used to extract image features. Finally, a fitting layer is used to determine the location of the lightning strike. The proposed method is simulated and implemented in MATLAB and Python, and the results show the effectiveness of the proposed method to determine the location of lightning strikes in a three-dimensional environment without requiring the use of at least three sensors. Manuscript profile

Open Access Article

15 - An Intrusion Detection System based on Deep Learning for CAN Bus
Fatemeh Asghariyan Mohsen Raji

In recent years, with the advancement of automotive electronics and the development of modern vehicles with the help of embedded systems and portable equipment, in-vehicle networks such as the controller area network (CAN) have faced new security risks. Since the CAN bu More

In recent years, with the advancement of automotive electronics and the development of modern vehicles with the help of embedded systems and portable equipment, in-vehicle networks such as the controller area network (CAN) have faced new security risks. Since the CAN bus lacks security systems such as authentication and encryption to deal with cyber-attacks, the need for an intrusion detection system to detect attacks on the CAN bus seem to be very necessary. In this paper, a deep adversarial neural network (DACNN) is proposed to detect various types of security intrusions in CAN buses. For this purpose, the DACNN method, which is an extension of the CNN method using adversarial learning, detects intrusion in three stages; In the first stage, CNN acts as a feature descriptor and the main features are extracted, and in the second stage, the discriminating classifier classifies these features and finally, the intrusion is detected using the adversarial learning. In order to show the efficiency of the proposed method, a real open source dataset was used in which the CAN network traffic on a real vehicle during message injection attacks is recorded on a real vehicle. The obtained results show that the proposed method performs better than other machine learning methods in terms of false negative rate and error rate, which is less than 0.1% for DoS and drive gear forgery attack and RPM forgery attack while this rate is less than 0.5% for fuzzy attack. Manuscript profile

Open Access Article

16 - Presenting a web recommender system for user nose pages using DBSCAN clustering algorithm and machine learning SVM method.
reza molaee fard Mohammad mosleh

Recommender systems can predict future user requests and then generate a list of the user's favorite pages. In other words, recommender systems can obtain an accurate profile of users' behavior and predict the page that the user will choose in the next move, which can s More

Recommender systems can predict future user requests and then generate a list of the user's favorite pages. In other words, recommender systems can obtain an accurate profile of users' behavior and predict the page that the user will choose in the next move, which can solve the problem of the cold start of the system and improve the quality of the search. In this research, a new method is presented in order to improve recommender systems in the field of the web, which uses the DBSCAN clustering algorithm to cluster data, and this algorithm obtained an efficiency score of 99%. Then, using the Page rank algorithm, the user's favorite pages are weighted. Then, using the SVM method, we categorize the data and give the user a combined recommender system to generate predictions, and finally, this recommender system will provide the user with a list of pages that may be of interest to the user. The evaluation of the results of the research indicated that the use of this proposed method can achieve a score of 95% in the recall section and a score of 99% in the accuracy section, which proves that this recommender system can reach more than 90%. It detects the user's intended pages correctly and solves the weaknesses of other previous systems to a large extent. Manuscript profile

Open Access Article

17 - Platform for manufacturing and intelligent production of polymers: genome engineering of polymer materials
Zeinab Sadat Hosseini

High-performance polymer materials are the foundation of high-level technology development and advanced manufacturing. Recently, polymeric material genome engineering (PMGE) has been proposed as a basic platform for the intelligent production of polymeric materials. Po More

High-performance polymer materials are the foundation of high-level technology development and advanced manufacturing. Recently, polymeric material genome engineering (PMGE) has been proposed as a basic platform for the intelligent production of polymeric materials. Polymeric Material Genome Engineering (PMGE) is an emerging field that combines the principles of the Materials Genome Initiative with polymer science to accelerate the discovery and development of new polymeric materials. The concept of PMGE is to create a comprehensive database of polymer properties obtained from both computational and experimental methods. This database can then be used to train machine learning models that can predict the properties of new polymers. However, the development of PMGE is still in its infancy and many issues remain to be addressed. Overall, PMGE represents a significant step towards the intelligent manufacturing of polymeric materials, with the potential to revolutionize the field by enabling faster and more efficient development of new materials. In this review are presented the fundamental concepts of PMGE and a summary of recent research and achievements, then are investigated the most important challenges and the future prospects. Specifically, this study focuses on the property prediction approaches, including of the proxy approach and the machine learning, and discusses the potential applications of PMGE, i.e. the advanced composites, the polymer materials used in the communication systems, and electrical integrated circuit manufacturing. Manuscript profile

Open Access Article

18 - Estimating the shear sonic log using machine learning methods, and comparing it with the obtained data from the core
Houshang Mehrabi Ebrahim Sfidari Seyedeh Sepideh Mirrabie Sadegh Barati Boldaji Seyed Mohammad Zamanzadeh

Machine learning methods are widely used today to estimate petrophysical data. In this study, an attempt has been made to calculate shear sonic log (DTS) from other petrophysical data using machine learning methods and compare it with the sonic data obtained from the More

Machine learning methods are widely used today to estimate petrophysical data. In this study, an attempt has been made to calculate shear sonic log (DTS) from other petrophysical data using machine learning methods and compare it with the sonic data obtained from the core. For this purpose, computational methods such as Standard Deviation, Isolation Forest, Min. Covariance, and Outlier Factors were used to normalize the data and were compared. Given the amount of missing data and box plots, the Standard Deviation method was selected for normalization. The machine learning methods used include Random Forest, Multiple Regression, Boosted Regression, Support Vector Regression, K-Nearest Neighbor, and MLP Regressor. Multiple regression had the lowest evaluation index (R2=0.94), while Random Forest regression had the highest correlation between the estimated shear sonic log and the original shear sonic log with an evaluation index of 0.98. Therefore, Random Forest regression was used for the final estimation, and to prevent data generalization or overfitting, the GridSearchCV function was used to calculate optimal hyperparameters and final estimation. The estimated sonic log showed a very high similarity with the core data. Manuscript profile

Open Access Article

19 - Intrusion Detection Based on Cooperation on the Permissioned Blockchain Platform in the Internet of Things Using Machine Learning
Mohammad Mahdi Abdian majid ghayori Seyed Ahmad Eftekhari

Intrusion detection systems seek to realize several objectives, such as increasing the true detection rate, reducing the detection time, reducing the computational load, and preserving the resulting logs in such a way that they cannot be manipulated or deleted by unauth More

Intrusion detection systems seek to realize several objectives, such as increasing the true detection rate, reducing the detection time, reducing the computational load, and preserving the resulting logs in such a way that they cannot be manipulated or deleted by unauthorized people. Therefore, this study seeks to solve the challenges by benefiting from the advantages of blockchain technology, its durability, and relying on IDS architecture based on multi-node cooperation. The proposed model is an intrusion detection engine based on the decision tree algorithm implemented in the nodes of the architecture. The architecture consists of several connected nodes on the blockchain platform. The resulting model and logs are stored on the blockchain platform and cannot be manipulated. In addition to the benefits of using blockchain, reduced occupied memory, the speed, and time of transactions are also improved by blockchain. In this research, several evaluation models have been designed for single-node and multi-node architectures on the blockchain platform. Finally, proof of architecture, possible threats to architecture, and defensive ways are explained. The most important advantages of the proposed scheme are the elimination of the single point of failure, maintaining trust between nodes, and ensuring the integrity of the model, and discovered logs. Manuscript profile

Open Access Article

20 - Application Identification Through Intelligent Traffic Classification
Shaghayegh Naderi

Traffic classification and analysis is one of the big challenges in the field of data mining and machine learning, which plays an important role in providing security, quality assurance and network management. Today, a large amount of transmission traffic in the network More

Traffic classification and analysis is one of the big challenges in the field of data mining and machine learning, which plays an important role in providing security, quality assurance and network management. Today, a large amount of transmission traffic in the network is encrypted by secure communication protocols such as HTTPS. Encrypted traffic reduces the possibility of monitoring and detecting suspicious and malicious traffic in communication infrastructures (instead of increased security and privacy of the user) and its classification is a difficult task without decoding network communications, because the payload information is lost, and only the header information (which is encrypted too in new versions of network communication protocols such as TLS1.03) is accessible. Therefore, the old approaches of traffic analysis, such as various methods based on port and payload, have lost their efficiency, and new approaches based on artificial intelligence and machine learning are used in cryptographic traffic analysis. In this article, after reviewing the traffic analysis methods, an operational architectural framework for intelligent traffic analysis and classification has been designed. Then, an intelligent model for Traffic Classification and Application Identification is presented and evaluated using machine learning methods on Kaggle141. The obtained results show that the random forest model, in addition to high interpretability compared to deep learning methods, has been able to provide high accuracy in traffic classification (95% and 97%) compared to other machine learning methods. Finally, tips and suggestions about using machine learning methods in the operational field of traffic classification have been provided. Manuscript profile

List of Articles یادگیری ماشین

Rimag

Links

Related Centers

Technical Support

Official pages